Reliability inference for a multicomponent stress–strength model based on Kumaraswamy distribution

In this paper, inference for a multicomponent stress–strength (MSS) model is studied under censored data. When both latent strength and stress random variables follow Kumaraswamy distributions with common shape parameters, the maximum likelihood estimate of MSS reliability is established and associated approximate confidence interval is constructed using the asymptotic distribution theory and delta method. Moreover, pivotal quantities based generalized point and confidence interval estimates are presented for the MSS reliability. Furthermore, likelihood and generalized pivotal based estimates are also presented when the strength and stress variables have unequal shape parameters. For complementary and comparison, bootstrap confidence intervals are provided as well under common and unequal parameter cases. In addition, to compare the equivalence between strength and stress shape parameters, the likelihood ratio test for hypothesis of interest is also discussed. Finally, simulation study and a real data example are provided to investigate the performance of proposed procedures. •Multicomponent stress–strength model is considered for latent failure times.•Existence and uniqueness of maximum likelihood estimators are established.•Pivotal quantities based generalized estimates are constructed.•Bootstrap percentile confidence intervals are also provided for comparison.•California Shasta reservoir water capacity data is investigated.